ASW2: "Integration of Functional and Taxonomic Diversity." October 2003 St Louis, MO, USA

Transcription

1 ASW2: "Integration of Functional and Taxonomic Diversity." October 2003 St Louis, MO, USA

2 What is Bio. M. E. R. G. E.? Biotic Mechanisms of Ecosystem Regulation in the Global Environment In a nutshell: Bring together those that study the diversity of our biota with those that study its functioning.

3

4 What is Bio. M. E. R. G. E.? 5 year Research Coordinating Network Funded by NSF Some co-functions with GCTE and DIVERSITAS 20 core participants > 60 general participants Inclusive: open to all interested parties Mechanism for interaction: Adaptive Synthesis Workshops

5 BioMERGE Mission Given just two things: 1. An inventory of species 2. Geography Can one state what the likely significance of biodiversity is to the region s ecosystem functions?

6 Results from ASW1

7 Policy is made even in the absence of scientific information We know human accelerated environmental change affects biodiversity We know biodiversity affects ecosystem functioning How do we effectively synthesize and communicate this knowledge Sala et al. 2000

8 Objective for ASW2 Translate i.e. Begin process of translating conceptual model into useful output using existing data on species distributions, functional traits and diversity/ecosystem function relationships

9 Goals for ASW 2 Identify 3-4 data sets where we have data on biota, their effect and response traits, and measures of ecosystem functioning (may require synthesis of large biotic inventories with trait data) Derive effect and response algorithms for these data sets Predict new levels of ecosystem function given addition of a driver to the system (how much does biodiversity matter?) Critically evaluate of proxies/shortcuts (e.g. FD, functional groups, scaling relationships) Identify key gaps in our data/model

10 Challenges for ASW2 How far can we get using readily available data? What data are necessary to fully implement model? Can we get these data? If not are there ways around this? Can we produce useful product?

11 Workshop Structure Aquatic Katia Engelhardt Martin Solan Brad Cardinale Jonathan Chase Amy Downing Jennifer Ruesink Diane Srivastava Jason Bradford Rob Colwell Ivette Perfecto Mahesh Sankaran Tropical Oliver Phillips Po Garden Peter Raven Jim Solomon Grasslands Andy Hector Clarence Lehman Jason Fridley David Hooper Jennie McLaren Amy Symstad Natalia Perez-Harguindeguy Justin Wright Theoretical Shahid Naeem Sandra Diaz Jennifer Hughes Claire Jouseau Sandra Lavorel Peter Morin Owen Petchey

12 Aquatic Group: Data Dataset on benthic community of Galway Bay, Ireland. 2 sites, many sampling dates Functional classification of species based on body size, burrowing depth, reworking mode Response: Biotic Mixing Depth (BMD) Sediment profiles from unpolluted (left) and polluted (right) sites in Galway Bay. Note shallower depth of BMD in polluted site due to lower functional diversity of bioturbators. Photos and data courtesy of M. Solan

13 Aquatic Group: Progress Developed relationship between community bioturbation potential (BP) and Biological Mixing Depth (BMD) Explored effects of biodiversity loss via Random species loss Body size model (largest species first) Abundance model (rarest species first) Pollution model (sensitive species first)

14 Aquatic Group: Next Steps Allow for compensation within and among functional groups following extinction Couple benthic and pelagic food webs Increase scale by using same analysis on data from Chesapeake Bay that has greater spatial and temporal extent

15 Grasslands Group: Data Data primarily from Cedar Creek (MN) BioCON experiment Measurements of many functional traits on most species in experiment made by Craine et al 2002 Response: Above-, belowground productivity, N retention, % transmittance. Can use N and CO 2 treatments to examine effects of changing drivers. Craine et al Functional traits, productivity and effects on nitrogen cycling of 33 grassland species. Functional Ecology 16:

16 Grasslands Group: Progress Developed 3 models of Functional Effects Algorithms (FEAs) 1. Determine EF of each species in monoculture. EF of mix = biomass weighted sum of EFs of constituent species 2. Incorporates species interactions: EF of mix = determined by EF of constituent species in mix rather than in monoculture 3. Incorporates trait density measures i.e. traits where dispersion of values is more important than mean of values (e.g. rooting depth and phenology) Traits identified as important in regulating ecosystem function in large-scale studies (e.g. SLA) not significant predictor at local scale Hypothesis environment selects for minimal variation in key traits that are important at broad scale. Other traits are more important at controlling small-scale variability in ecosystem functioning.

17 Grasslands Group: Next Steps Clean up data from BioCON experiment and fill in functional trait holes Develop appropriate trait density statistics Select traits for model based on mechanistic assumptions Compare 3 FEA models & success of functional traits selected based on local analysis vs. large-scale analysis Potential future papers looking at FEAs when confronted with altered drivers; look across ecosystems to determine what functional traits are important for determining ecosystem functioning

18 Tropical Group: Data RAINFOR data set: 1 ha plots surveyed 9-23 years all individuals mapped and measured Traits: wood density (taxonomic), pioneer index, cation association, moisture association, N- fixer (taxonomic) Response: Changes in distribution/composition & biomass accrual

19 Tropical Group: Progress Integration of trait data base Recognition of orthogonal functional traits: pioneer index (measured by wood density) and size (measured by basal area) Used long-term data to estimate how increases in CO 2 affect distribution of functional traits small decrease in median wood density

20 Tropical Group: Next Steps More refined trait measurements (e.g. wood density for all species) Estimate standing forest biomass and turnover rates given changes in distribution of functional traits Incorporate additional drivers (wood harvesting) More efficient database design to allow for further analyses Incorporate stem data from more recent surveys

21 Theoretical Group: Principles A fundamental theoretical BEF framework founded on basic ecological principles would serve to integrate findings from the case studies (aquatic, grassland, and tropical) A theoretical framework needs to integrate principles from Community ecology Eltonian pyramid Trophic structure/dynamics Scaling rules The niche The biotope Ecosystem ecology Abiotic constraints on ecosystem/biogeochemical processes Coupled patterns of material cycling and energy flow Biogeography Log-normalish patterns of distribution and abundance Gradient effects Island effects

22 Theoretical Group: Progress We developed a theoretical BEF framework that can be examined by simulation We developed an algorithm for constructing the biogeographic matrix that creates a biota consistent with biogeographical principles We developed an algorithm for constructing the responseeffect matrix that structures the biota consistent with community and ecosystem principles

23 Theoretical Group: Next Steps Conduct simulations and compare findings with current empirical evidence Refine framework and improve precision Restructure framework for grasslands, tropical, and estuarine systems Simulate four scenarios Type I. Invasion. New species are added from other ecosystems. Type II. Over-exploitation. Dominant species are overexploited. Type III. Random extinction. No particular driver is considered. Type IV. Forced. Increasing eutrophication.

24 ASW2 Conclusions Significant progress made in translating measures of biodiversity and functional traits into ecosystem function in some ecosystems Generally, datasets that allow for this are rare Need more complete measures of functional traits of species Need better measurements of role of functional diversity in controlling ecosystem functioning Significant progress was made in refining the theoretical framework for BEF Need to test framework by model building and simulation